embulk-plugin-input-hbase 0.0.2-java

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 129327d9138c961f83ee26a91824f0022612bff1
+  data.tar.gz: a3312beeaee9a9f5da3a62284b41c7829095ca57
+SHA512:
+  metadata.gz: a903a3d04fae3a12e2e77be45be3b2a9dbc4c6cd7daf457c131e430d03b9660fe6b5e3a7d95bdeaaf3cf7d0061b88d40ec29ce397706efa373353a12a0383e29
+  data.tar.gz: 4e3a371d64df8228199471d719ebb47bba4e2b13dd11a48a2653c3625b098bfff050ee1b64dfc460e10acd976ce736b2eebde009d4d7529cdc02b8c3bb6e8d6b

data/README.md

@@ -0,0 +1,42 @@
+# embulk-plugin-input-hbase
+
+## Example
+
+HBase table:
+
+```ruby
+hbase(main):029:0> scan 'example:test'
+ROW                                  COLUMN+CELL                                                                                               
+ r1                                  column=foo:dig, timestamp=1422458241976, value=\x00\x00\x00\x00\x00\x00\x00\x01                           
+ r2                                  column=foo:dig, timestamp=1422458257028, value=\x00\x00\x00\x00\x00\x00\x00\x02                           
+ r2                                  column=foo:str, timestamp=1422458830978, value=hello                                                      
+ r3                                  column=foo:str, timestamp=1422458270762, value=hey                                                        
+3 row(s) in 0.0860 seconds
+```
+
+Embulk config:
+
+```yaml
+in:
+  type: hbase
+  host: localhost
+  table: 'example:test'
+  columns:
+    - {name: 'foo:dig', type: long}
+    - {name: 'foo:str', type: string}
+out:
+  type: stdout
+```
+
+Embulk preview:
+
+```bash
+$ java -jar embulk.jar preview -C $(hbase classpath) example.yml
++--------------+----------------+
+| foo:dig:long | foo:str:string |
++--------------+----------------+
+|            1 |                |
+|            2 |          hello |
+|              |            hey |
++--------------+----------------+
+```

data/lib/embulk/input_hbase.rb

@@ -0,0 +1,67 @@
+require 'java'
+
+java_import org.apache.hadoop.hbase.HBaseConfiguration
+java_import org.apache.hadoop.hbase.client.HConnectionManager
+java_import org.apache.hadoop.hbase.client.Scan
+java_import org.apache.hadoop.hbase.util.Bytes
+java_import org.apache.hadoop.hbase.CellUtil
+
+module Embulk
+  class InputHBase < InputPlugin
+    Plugin.register_input('hbase', self)
+
+    def self.transaction(config, &control)
+      task = {
+        'host' => config.param('host', :string, default: 'localhost'),
+        'table' => config.param('table', :string)
+      }
+      threads = 1
+      columns = config.param('columns', :array).map.with_index { |column, i|
+        Column.new(i, column['name'], column['type'].to_sym)
+      }
+      commit_reports = yield(task, columns, threads)
+      return {}
+    end
+
+    def initialize(task, schema, index, page_builder)
+      super
+    end
+
+    def run
+      conf = HBaseConfiguration.create
+      conf.set('hbase.zookeeper.quorum', @task['host'])
+      connection = HConnectionManager.createConnection(conf)
+      table = connection.getTable(@task['table'])
+      scan = Scan.new
+      scanner = table.getScanner(scan)
+      scanner.each { |result|
+        @page_builder.add(@schema.map { |column|
+          family, qualifier = column.name.split(':').map {|e|
+            Bytes.toBytes(e)
+          }
+          raw = nil
+          if result.containsColumn(family, qualifier) then
+            cell = result.getColumnLatestCell(family, qualifier)
+            raw = CellUtil.cloneValue(cell)
+          end
+          if raw then
+            case column.type
+            when :long
+              Bytes.toLong(raw)
+            when :string
+              Bytes.toString(raw)
+            else
+              raw
+            end
+          else
+            nil
+          end
+        })
+      }
+      @page_builder.finish
+      commit_report = {
+      }
+      return commit_report
+    end
+  end
+end

metadata

@@ -0,0 +1,73 @@
+--- !ruby/object:Gem::Specification
+name: embulk-plugin-input-hbase
+version: !ruby/object:Gem::Version
+  version: 0.0.2
+platform: java
+authors:
+- Shun Takebayashi
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2015-02-04 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  name: bundler
+  prerelease: false
+  type: :development
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.0'
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: 0.9.2
+  name: rake
+  prerelease: false
+  type: :development
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: 0.9.2
+description: Embulk plugin for HBase input
+email: shun@takebayashi.asia
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/embulk/input_hbase.rb
+- README.md
+homepage: https://github.com/takebayashi/embulk-plugin-input-hbase
+licenses:
+- Apache 2.0
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project:
+rubygems_version: 2.1.9
+signing_key:
+specification_version: 4
+summary: Embulk plugin for HBase input
+test_files: []